GLACIERS OF TJNAKWTK INLET AND COLLEGE FIORD 291 nearly as far north as Yale Glacier instead of at a point nearly east of Crescent Glacier in College Fiord. The National Geographic Society's expedition of 1910 spent July 12th to 14th in the study of Meares Glacier and Unakwik Inlet, also making the hachure map of the glacier reproduced as Fig. 37 and the soundings shown on Fig. S8.1 Grant has stated 2 that in 1905 "the bushes and trees were close to the ice and there was no bare zone, or at most a very narrow one, visible between the ice and the forest. In 1909 the front of the ice seemed to be a little in advance of its position of four years before. At the later date near the front of the glacier on the south, side was a brown zone estimated to be 200 feet in width.. This brown zone appeared to have been caused by dead vegetation rather than by bare rock, and at the edge of the ice there were a few small trees. Close to the glacier there was a sparse forest which contained trees estimated to be ten inches in diameter. Hence the ice was probably as far forward in 1909 as it has been during the last hundred years." In 1910 the north edge of the glacier (PI. CXUI) had advanced slightly since the year before, as was shown by a comparison of photographs and by the nearly vertical terminal ice cliff on the beach, at the base of which were freshly-fallen ice blocks. Near the southern side we were unable to determine positively whether the glacier had remained stationary from 1909 to 1910, or whether there had been a slight retreat. A mass of ice, dinging to the. mountain side at sea level, was connected with the glacier in 1909, but by July, 1910, had been detached by melting. It is easy to understand the advance of one side of this glacier and the stationary condition or slight retreat of the other, in view of the fact that the tributary from the east is relatively inactive, while the one from the north was more active in 1910 than in the year before. This difference in activity of the tributaries is shown by the conditions along the margins of the glacier below their junction. On the south margin there was no change in condition during the last two years of observation, while along the north margin the glacier was spreading as well as advancing, everywhere extending up to the forest and uprooting the turf, in places pushing back into and even overriding the forest. These conditions were observed both at the terminus and up to the turn in the glacier. Slivers of ice projected among the alders and spruces and small push moraines of till and turf were seen. Several spruces, not yet overturned, stood at the very ice margin, here stained dark colored with lateral moraine. In the quotation given above, Grant refers to a brown zone of dead vegetation on the south margin of the glacier in 1909, and tTiia was not overridden in 1910. It may be interpreted as the result of forest destruction during an advance of the east tributary before 1909. One mature tree, tilted during the advance, stood at the edge of this brown zone and there were several young shrubs at the very edge of the ice. The terminus of the glacier was precipitous in 1910, as if the end were in deep water, and at a distance of a half mile the fiord is 584 feet deep, suggesting that the end of the glacier is not afloat. We agree with Grant that Meares Glacier is probably as far advanced now as it has been in the last century. There is considerable forest on the mountain slope above the i Martin, Lawrence, The National Geographic Society Researches in Alaska, Nat. Geog. Mag., Vol. 22, 1011, p. 551. • Bull. Amer. Geog. Soc., Vol. XLH, 1910. p. 738.